seg fault with last git version

Dirk Hohndel dirk at hohndel.org
Sun Nov 2 10:54:00 PST 2014


On November 2, 2014 10:43:54 AM Pedro Neves <nevesdiver at gmail.com> wrote:

> HI:
>
> On 11/02/2014 06:36 PM, Dirk Hohndel wrote:
> > sorry, missed a character. I meant
> >
> > p *dc
>
> (gdb) p *dc
> $2 = {when = 1156089600, duration = {seconds = 3000}, surfacetime =
> {seconds = 0}, maxdepth = {mm = 41000}, meandepth = {mm = 0}, airtemp =
> {mkelvin = 0}, watertemp = {mkelvin = 288050}, surface_pressure = {mbar
> = 0}, dctype = OC, no_o2sensors = 0 '\000', salinity = 0,
>    model = 0x21f0c5e0 "ADL Conduit", deviceid = 4294967295, diveid = 0,
> samples = 0, alloc_samples = 0, sample = 0x0, events = 0x21f0c600, next
> = 0x0}
> (gdb)

OK, as I guessed. Fix was just pushed to master.

> > my guess is that no samples are allocated, so peeking at sample[0] fails.
> >
> > Can you send us the XML file that you are loading? In the meantime I’ll 
> commit a quick fix that ensures that we don’t SEGV here.
>
> My log is 28.6 Mb in size... Maybe I can have a look at the file, if I
> can get some pointers on where to look...

How was the dive with the ADL Conduit downloaded? Seems like this is from 
August 2006 if my math is right (not at my computer right now)

/D






More information about the subsurface mailing list